# -----------------------------------------------------------------------------
# Immo-Ka — Agrégateur de maisons à vendre (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/m_immobilier.py : M Immobilier (mimmobilier.com)
#   Agence indépendante de prestige (Grand Montréal), inscriptions Centris.
#   La page /properties est rendue serveur : chaque carte « cardProperty »
#   porte l'URL (avec no Centris), la ville, l'adresse, chambres, salles de
#   bains, prix et les photos (/images/centris-slideshow/{id}-N-*.jpg).
# -----------------------------------------------------------------------------
from __future__ import annotations

import html as _html
import os
import re

from .base import BaseConnector
from . import _detailutil as du
from ..normalize import parse_price
from ..schema import PropertyListing

BASE = "https://www.mimmobilier.com"
LISTING_URL = f"{BASE}/properties"
CARD_SPLIT = "cardProperty col-span-12"
HREF_RE = re.compile(r'href="(/properties/[^"]+/(\d+))"')
IMG_RE = re.compile(r'(/images/centris-slideshow/\d+-\d+-\d+\.(?:jpg|jpeg|png|webp))', re.I)
DETAIL_LIMIT = int(os.environ.get("IMMOKA_M_DETAIL_LIMIT", "300"))


class MImmobilierConnector(BaseConnector):
    source_id = "m_immobilier"
    request_delay = 0.6
    use_detail_cache = False

    def fetch(self) -> list[PropertyListing]:
        html = self.get(LISTING_URL).text
        cards = html.split(CARD_SPLIT)[1:]
        out: dict[str, PropertyListing] = {}
        for card in cards:
            card = CARD_SPLIT + card[:6000]
            lst = self._parse_card(card)
            if lst and lst.uid not in out:
                out[lst.uid] = lst
        listings = list(out.values())
        # fiche détail : galerie complète (135 photos) + description + type
        du.enrich(self, listings, DETAIL_LIMIT, parse_m_detail, key="v2")
        return listings

    def _parse_card(self, card: str) -> PropertyListing | None:
        m = HREF_RE.search(card)
        if not m:
            return None
        url = BASE + m.group(1)
        external_id = m.group(2)

        images = []
        for im in IMG_RE.findall(card):
            full = BASE + im
            if full not in images:
                images.append(full)

        # texte du carton, ligne par ligne
        text = _html.unescape(re.sub(r"<[^>]+>", "\n", card))
        lines = [l.strip() for l in text.splitlines() if l.strip()]

        status_txt = lines[1] if len(lines) > 1 else ""
        # ville et adresse : les 2 lignes après le statut « À vendre »
        city = address = ""
        try:
            k = next(i for i, l in enumerate(lines)
                     if l.lower().startswith(("à vendre", "a vendre", "vendu")))
            status_txt = lines[k]
            city = lines[k + 1] if k + 1 < len(lines) else ""
            address = lines[k + 2] if k + 2 < len(lines) else ""
        except StopIteration:
            pass

        price_label = _after(lines, "prix")
        if price_label and "/ m" in price_label.lower():
            return None                    # location, pas une vente
        price = parse_price(price_label or "")
        beds = _int(_after(lines, "chambres"))
        baths = _int(_after(lines, "salles de bains"))

        vendu = "vendu" in status_txt.lower()
        return PropertyListing(
            source=self.source_id,
            external_id=external_id,
            url=url,
            title=f"{address}, {city}".strip(", "),
            address=address,
            city=city,
            property_type="",              # non exposé sur le carton
            price=price,
            price_label=price_label or "",
            bedrooms=beds,
            bathrooms=baths,
            mls=external_id,
            status="vendu" if vendu else "a-vendre",
            images=images,
            broker_name="M Immobilier",
        )


def parse_m_detail(html: str) -> dict:
    """Galerie Centris complète + description (JSON-LD) + caractéristiques."""
    out: dict = {}
    imgs = []
    seen = set()
    for im in IMG_RE.findall(html):
        full = BASE + im
        if full not in seen:
            seen.add(full)
            imgs.append(full)
    if imgs:
        out["images"] = imgs
    desc = du.ld_description(html)
    if desc:
        out["description"] = desc
    _det = du.centris_details(du.flatten(html))
    if _det:
        out.setdefault("details", {}).update(_det)
    return out


def _after(lines: list[str], label: str) -> str:
    lab = label.lower()
    for i, l in enumerate(lines):
        if l.lower() == lab and i + 1 < len(lines):
            return lines[i + 1]
    return ""


def _int(s: str):
    if not s:
        return None
    # « 4 + 1 » -> 5 (chambres principales + sous-sol)
    nums = [int(x) for x in re.findall(r"\d+", s)]
    return sum(nums) if nums else None